@dambuildshit

I build apps and ship fast. Documenting the journey from 0. Day 1. 🚀

Joined February 2026
  • Tweets225
  • Following33
  • Followers21
52Media

My 4-month-old AI site is about to overtake a 10-year-old domain. Here's the data.

Article · @dambuildshit
1
1
1
151
Is it just me, but I find Opus 5.5 in the desktop app look like more opus 6. In the cli, it feel less more powerful. Just me?
13
What's going on with gpt-4o cult?
45
I can smell new GPT model dropping soon, when Astra behave like a junior Pakistanese developer from fiverr
2
This morning I have the surprise to see that Astra decided to ssh into one of my prod server, and dump the prod db to launch the local app. I just told him to launch the local app, but it might have think that it was better to do it with the prod datas...unbelievable.
1
2
wow, then it's flagged by OpenAI
1
I have 50% remaining on my Claude subscription for 2 days, and an app idea. Let make it happen. See you Sunday.
1
1
1
156
Woke up, Claude and his crews worked quite well.
1
9
The entire app is almost complete. Claude still running. 8h run so far and counting. Althought limit is coming soon, so I guess. I'll have to wait tomorrow for the full completion. Impressive.
4
I'm blown away that Google put such a useless Cloud Assistant... Not only GCP require some nerves and patience to navigate. But the AI assistant does not even use their most up to date model. @googlecloud
10
Most of my apps have been on the OpenAI API because Claude’s API was usually too expensive to justify. Anthropic just rolled out monthly Claude Platform credits for Max and Team plans , so I moved two of them over. Let's see how they hold up in production. Who else is making the shift?
1
17
One client app used to take 42 minutes to fully deploy. After some work with Astra, it’s down to 16 minutes. Image build and publish: 7 min 15 s → 2 min 52 s. Total cut by about 63%.
1
1
12
Started to make an old school shoot them up with GPT 6.1. Thanks to CC, Codex, Cursor, I'm able to ship client project faster, and spend time doing some cool shit. Exciting time.
1
17
Been doing doing some improvement: The shot is the real fun part. Grab a power-up and the plane’s fire changes, then gets stronger again on the next one. First two levels in the clip. Went from Pong-shaped to an actual old-school shoot ’em up in less than a day. Three boss levels, still rough.
12
Opus 5.5 is really good. Have been doing a lot of long run work with it. I have it creating a full dashboard and force subagent to keep track for me, and them on what's implemented. And with a good prompt, it run for hours, one shooting heavy refactoring. Verdict in couple of hours, but I'd bet on 2-3hours run.
1
1
27
I guess it will run longer...
6
Rule I always use with AI APIs: load $50–100, turn auto top-up off, one project one key. Today an agent update went into a retry loop and ate $50. Fine. If auto top-up was on that same loop could’ve been hundreds. Small prepaid balance is the cheapest insurance you can buy.
8
Spent the last 3 hours finishing a prod server setup for a client workflow: Codex → Cmd+C Claude Code on the server → Cmd+V repeat I should’ve just given Codex SSH access and let it cook instead I became the world’s most expensive clipboard between two AIs client still thinks this is devops
1
1
40
Tried DeepSeek v4.1 Flash after all the X hype. To me it's a mixed bag. It’s fast, definitely. But it burns a lot of tokens, and the implementation I built on it came out with many issues. Curious if that’s just me or if a lot of people are landing in the same place.
1
24
Almost 1 billion tokens spent on DeepSeek v4.1 flash, Still not convince. What I'm missing?
25
Spent ~1h with Astra speccing a full admin panel for one of my app. Then launched a run with specific rules: - Astra: don’t write code - only review slices from 5.6 Terra - definition of done = ready to deploy 7h 18m later. 40% of my weekly limit gone. It’s “done.” A complete admin panel. On dummy data. 👀
1
1
296
Load more